Using Large Language Models and Knowledge Graphs to Improve the Interpretability of ML Results in Manufacturing

Forum topic · 小凯 · 2026-04-21

Summary

A 2026 arXiv paper (2604.16280) by Bayer, Lohr, Weiß, Michelberger, and Höpken proposes a method to improve the explainability of Machine Learning (ML) results by combining Knowledge Graphs (KGs) with Large Language Models (LLMs). Domain-specific data, ML results, and their explanations are stored in a knowledge graph, creating a structured link between domain knowledge and ML insights. A selective retrieval approach extracts relevant triplets from the KG, which are then processed by an LLM to produce user-friendly explanations. The method was evaluated in a manufacturing setting using the XAI Question Bank, covering 33 questions including custom, more complex ones. Answers were analyzed with quantitative metrics (accuracy, consistency) and qualitative metrics (clarity, usefulness). The work contributes theoretically, by enabling LLMs to dynamically access KG content for explaining ML results, and practically, by providing empirical evidence that such explanations work in real manufacturing environments and support better decision-making in manufacturing processes.

Paper Overview

Field: Machine Learning
Authors: Thomas Bayer, Alexander Lohr, Sarah Weiß, Bernd Michelberger, Wolfram Höpken
Published: 2026-04-17
arXiv: 2604.16280

Original Abstract

Explaining Machine Learning (ML) results in a transparent and user-friendly manner remains a challenging task of Explainable Artificial Intelligence (XAI). In this paper, we present a method to enhance the interpretability of ML models by using a Knowledge Graph (KG). We store domain-specific data along with ML results and their corresponding explanations, establishing a structured connection between domain knowledge and ML insights. To make these insights accessible to users, we designed a selective retrieval method in which relevant triplets are extracted from the KG and processed by a Large Language Model (LLM) to generate user-friendly explanations of ML results.

We evaluated our method in a manufacturing environment using the XAI Question Bank. Beyond standard questions, we introduce custom, more complex questions that highlight the advantages of our approach. We evaluated 33 questions, analyzing the generated answers with quantitative metrics (accuracy and consistency) as well as qualitative metrics (clarity and usefulness).

Contributions

1. Theoretical: A novel method that enables LLMs to dynamically access KG content, improving the interpretability of ML results. 2. Practical: Empirical evidence that LLM-generated, KG-grounded explanations can be successfully applied in real manufacturing environments, supporting better decision-making in manufacturing processes.

---

*Auto-collected on 2026-04-21.*